The proposed project takes an innovative approach to childhood obesity prevention, for which there currently no evidence-based programs, and for which results of current trials have produced mainly short-term or disappointing effects. The aim of this project is to adapt and revise parts of two nationally recognized evidence-based programs for drug prevention (the school program PATHS, and the parent skills training program STAR) for use with children in grades 4, 5, and 6 with the express purpose of obesity prevention. To the investigators'knowledge, there is no school-based obesity prevention program that translates methods from other prevention areas, incorporates affect (specifically impulse control and emotional regulation) and executive function (planning and organization), and addresses social competence-building (a major developmental task of these childhood years) as a means to prevent obesity. A total of 24 elementary schools from two of the largest districts in Orange County will be randomly assigned to either the universal obesity prevention program or a delayed intervention control (N=3460 4th grade students and their parents). The adaptation is planned as a three year program with boosters, administered by trained teachers. The population is ethnically diverse (36% white, 57% Hispanic, 6% Asian;48% on free/reduced lunch programs). Self-report measures will be administered on the longitudinal panel sample at baseline in Fall of 4th grade, and at end of 4th, 5th, and 6th grade on cognitive, affective, and behavioral variables related to the intervention;anthropomorphic measures of BMI and waist circumference will be administered at the beginning of 4th grade, and at end of 5th and 6th grade. Teacher, administrative, and parent surveys will be administered on the same schedule to measure school environment for physical activity and food;child emotional regulation, social competence, executive function, and internalizing and externalizing behavior, and parent-child communication. Program implementation will be measured by teacher self-report and research staff observations. Data will be analyzed with mixed model regression to capture effects of school and classroom, structural equation modeling to test the theoretical model of change, and limited growth curve analysis to evaluate developmental trends in mediators and outcomes across the three grades. Findings should be generalizable to most elementary schools, and will be used to develop evidence-based program standards for childhood obesity prevention.